1 Rev C

A STUDY COMPARING ELECTRIC, OIL, AND PRESSURIZED

WATER HEATING FOR COMPOSITE MOLDING

Kip Petrykowski, Jim Fischer

Single Temperature Controls

14201 South Lakes Dr. Suite B

Charlotte, NC 28273

ABSTRACT Electric cartridge heaters have been used for many years to heat composite molds. They are

easy to install and operate and thus make a logical choice. Unfortunately, they consume large

amounts of electricity to operate, do not offer cooling, and are notorious for inconsistent

heating.

Oil heating has been used to heat and cool composite molds for an equally long period

primarily as a carry-over from its use in other plant equipment and plant heating. Unfortunately

it is very slow to build temperature, consumes large amounts of electricity, and does not fit

with the modern greener philosophy of manufacturing.

Due to the inherent energy savings, high temperature spectrum, precise temperature control,

and fast ramp rates, pressurized water offers numerous advantages over both systems when

applied to composite molding.

1. INTRODUCTION

As more OEs eliminate oil heating from clean room environments and utilities cost increase,

the need for a viable option to oil and electric cartridge heating has been created. Water due to

its inherent properties makes an excellent choice. With the use of pressurization to increase the

temperature spectrum, water can be a direct replacement for oil systems. Due to the inherent

energy savings, high temperature spectrum, precise temperature control, and fast ramp rates for

both heating and cooling, pressurized water systems have been used for injection molding for

over 35 years. There are also numerous Aerospace companies that have recently adopted it as

part of their Out-of-Autoclave (OoA) strategies.

1.1 Intent

The intent of the study and consequent desire to publish or present the data, was to help

provide molders with better insight into when to use each of the three (3) solutions for the

molding of composite parts. Publishing and presenting of the study should also help to

educate the Composites sector in technology that is not yet as ubiquitous as electric or oil

heating but provides for energy consumption reductions and more consistent part quality.

1.2 Audience

Companies or persons engaged in the manufacture of composite parts.

2 Rev C

1.3 Scope

The study is focused on (3) three primary heating methods; electric cartridge, pressurized

water, and oil. It does not attempt to compare Ovens against the three methods.

1.4 Background

Much of the motivation for the study was due to customer questions about cycle times, energy

consumption, temperature precision, mold temp gradient, and ramp rates they could expect as

they moved from oil and electric cartridge heaters to pressurized water systems. As a

manufacturer of temperature control units in both water and oil, Single did not typically get

involved in mold or mold design or the prediction of ramp rates for. In order to better answer

customer questions about these items, the study was commissioned.

The current industry perception is that oil is more precise for mold temperature control than

electric cartridge but slower. Cartridges are notorious for large temperature gradients across

molds but are relatively fast and easy to install. Pressurized water is relatively unknown to this

sector.

2. EXPERIMENTATION

2.1 Test parameters for Oil versus Pressurized water

1. Use equipment with similar flow rates through the mold. 2. Use equipment with similar heating/cooling capacities. 3. Use similar line sizes to the test mold from both test units.

Goal:

1. Determine Temperature Control Unit (TCU) temperature profiles, mold temperature profiles, and energy consumption for both systems.

Test Equipment:

Item Description

Used Single H0.2 12 kW Heating / 41 kW Cooling

Hours 10 Hours on unit

Power Supply 460 V/60 Hz (3 phase)

Flow 60 liters/minute rated flow

Heating Lines out to mold 10 mm I.D

Line Length 1.5 m

Number of Lines 2

Cooling lines into TCU 10 mm I.D

Line Length 3.5 m

Number of Lines 2

3 Rev C

Used Single D0.2 24 kW Heating / 116 kW Cooling

Hours 100 Hours on unit

Power Supply 460 V/60 Hz (3 phase)

Flow 100 liters/minute rated flow

Heating Lines out to mold 18 mm

Line Length 3.5 m

Number of Lines 2

Cooling lines into TCU 15 mm I.D

Line Length 3.5 m

Number of Lines 2

Mold Webber externally plumbed Nickel, single

sided test mold

Mold Weight 35.6 Kg

Measured flow through mold (water) 25 liters

Method Differential pressure

Measured flow through mold (oil) 28 liters

Method Differential pressure

Ambient air temperature for test 18.3C

Temperature probe Atkins Series 384 Digital

Amp Meter Fluke 442

Chilled water supply temperature 26.6C

Flow (water) (60 l/minute)

Figure 1.

Comments:

1. In order to use units from stock, an oil unit with 2x the heating capacity, 2.8x the cooling capacity, and 1.6x the maximum rated flow as compared to the water unit was used.

2. The oil units inside diameters for the heating/cooling inlets were 1.8x larger than the inside diameters for the water units heating/cooling inlets, thereby allowing for

significant gains in flow rates through the mold.

Experimental Procedure

A nickel shell externally plumbed mold was provided by Weber Manufacturing, Ontario,

Canada, for the test. The Weber mold was chosen because of its ability to be heat balanced

by controlling the location and number of tubes used to deliver the liquid medium.

Temperature readings confirmed that variation across the mold at temperature was within the

4 Rev C

desired

5 Rev C

0

50

100

150

200

250

0 10 20 30 40

Temperature

(C)

Time (Minutes)

Water TCU vs. Oil TCU Heating Profiles

Water TCU

Oil TCU

Figure 3.

Figure 4.

6 Rev C

Figure 5.

Figure 6.

7 Rev C

Figure 7.

8 Rev C

2.2 Test parameters for Electric Cartridge heaters versus Pressurized water

1. Use the same mold for both cartridge heaters and pressurized water. 2. Use the same size holes in the mold for both cartridge heaters and pressurized water. 3. Use equipment with similar heating capacities.

Goals:

1. Determine mold temperature profiles and energy consumption for both systems.

Test Equipment:

Item Description

Used Single H0.2 6 kW Heating / 41 kW Cooling

Hours 100 Hours on unit

Power Supply 460 V/60 Hz (3 phase)

Flow 60 liters/minute rated flow

Heating Lines out to mold 8 mm I.D

Line Length 1.5 m

Number of Lines 2

Cooling lines into TCU 8 mm I.D

Line Length 1.5 m

Number of Lines 4

Mold Two sided test mold

Mold Weight 18 Kg

Calculated flow through mold (water) 10 liters

Method Single Temp Heat Transfer Spreadsheet

Ambient air temperature for test 21C

Temperature probe Atkins Series 384 Digital

Amp Meter Fluke 442

4x 750 Watt 5/16" x 5" cartridge heaters 240V

Figure 8.

Comments:

1. The mold temperatures for the electric cartridge tests were controlled by comparison of

input received from a thermocouple placed in the mold to a temperature set point on the

units controller.

9 Rev C

2. The mold temperatures for the water tests were NOT controlled by comparison of input

received from a thermocouple placed in the mold to a temperature set point on the units

controller. The controller on the unit was set to the set point and allowed to monitor the

temperature via the incoming water temperature from the mold.

3. In order to allow the same mold to be used for both methods, the mold was not

optimized for the water unit. Limited flow due to the small mold passages would only a

flow of 9 liters/minute, thus not allowing for much energy to be transferred to the mold.

4. The unit, due to availability, was also the smallest model and thus had substantially less

available heating capacity than the cartridge heaters. The unit at 6kW had 1/5 the

heating capacity as compared to the electric cartridge heaters.

Experimental Procedure

Holes were drilled 19mm below both parts centered on a 50mm width on both the stationary

and moving halves of the mold. The diameter of these holes was properly sized for a 130mm

cartridge heater placed directly below the entire part. These same holes were also used as

water channels with the heater cartridges removed. The water was required to turn multiple

90 turns and exited the same end as it entered. Not typically a preferred mold design but

allowed for water and cartridge placement directly under the part. A thermocouple was

centered between the two plaques and on the centerline of the 130mm length on both mold

halves for mold temperature monitoring and for control with the electric heat.

10 Rev C

Figure 9.

Figure 10.

11 Rev C

The material used for all trials was polyetheretherketone (PEEK). The mold temperature

controller was set to 193C and the temperature reading from the thermocouple was allowed to

stabilize with the mold closed. Mold temperature readings were taken using a surface

pyrometer on the molding surfaces near the gate, in the part center, and opposite the gate after

heat soak after (1) one and (2) two hours of production. After the 2-hour runs for each part

were completed, the heater cartridges were removed and the mold was set-up using the

pressurized water temperature control unit. The same molding process established for the

electric heat was used for the pressurized water. Mold temperature readings were taken just as

was performed for the electric heat trials.

The consistency of the thermocouple reading was also noted during the molding trials. After

the mold reached steady state during the initial heat soak period, the mold was opened to

prepare for the trial. During the electrically heated trials, any changes to the steady state

condition yielded large temperature swings in the thermocouple reading. The thermocouple

reading changed by as much as +/- 13C in these trials. The thermocouple reading dropped

significantly at this time, but did recover to reach steady state at the set point within 10

minutes. The thermocouple temperature of the water heated molds varied by only +/- 5C and

was not greatly influenced by changes in the steady state operating condition.

The mold surface temperature was taken at 12 places in the beginning, middle, and

end of each (2) two hour run with a surface pyrometer.

12 Rev C

Heat

Type Part Time Temperature - C

Electric

2.54

mm

Thick

Soak

Temp. 163.3 173.3 129.4 158.3 175.6 130.6 157.2 169.4 149.4 160.6 172.8 150.0

1 Hour 154.4 165.0 142.2 150.6 165.0 147.2 157.2 173.3 158.3 158.3 171.7 156.1

2 Hours 156.1 168.9 142.8 155.6 168.3 142.8 155.6 172.2 160.6 156.1 168.9 157.8

5.08

mm

Thick

Soak

Temp. 158.3 169.4 135.6 155.0 170.0 141.1 154.4 174.4 145.0 159.4 177.2 147.2

1 Hour 153.3 165.0 141.1 156.1 171.1 143.3 156.7 173.9 158.9 164.4 178.9 161.7

2 Hours 155.0 166.7 141.1 158.3 176.1 145.0 155.0 171.1 157.8 163.9 177.2 159.4

Water

2.54

mm

Thick

Soak

Temp. 172.8 175.0 172.2 174.4 176.1 174.4 173.3 174.4 172.8 174.4 175.6 175.6

1 Hour 171.7 176.7 172.2 173.9 175.6 172.2 173.3 176.7 174.4 175.0 176.1 176.7

2 Hours 172.2 177.2 171.7 172.8 177.2 173.3 172.8 177.8 175.0 175.6 176.7 176.1

5.08

mm

Thick

Soak

Temp. 170.0 172.2 172.2 175.0 175.0 174.4 172.2 172.2 171.7 173.9 175.0 174.4

1 Hour 172.8 176.1 172.8 175.0 178.3 173.3 171.7 175.6 173.9 176.7 177.8 177.8

2 Hours 172.8 176.1 172.8 174.4 178.9 173.9 172.2 175.0 174.4 174.4 176.1 175.6

Location A B C D E F G H I J K L

Figure 11.

13 Rev C

Figure 12.

Figure 13.

14 Rev C

Figure 14.

A review of the temperature readings indicates that the water heated runs were more consistent

over the 2 hours. The average differential in temperature of the same location over the 2 hours

in the water heated runs was 1.94C, while the average differential in temperature of the same

location over the 2 hours in the electrically heated runs was 6.7C.

A more concerning point is the difference in temperature over the length of a part. The

average temperature differential over the length of one part at the same moment in time is

21.0C for the electrically heated trials. The average for the water heated trials was only

2.7C. This large temperature differential over the length of one part in the electrically heated

mold could create crystallinity changes within a part. This would create shrinkage differences

throughout the part/ resulting in a high degree of molded in stress. This stress could cause part

warpage, creep, and physical property differences within the same part. In order to confirm an

acceptable level of crystallinity the parts were analyzed with differential scanning calorimetry

(DSC) to determine the % crystallinity in the parts.

The DSC results confirm the concerns raised by the mold temperature readings during the

molding trials and the part dimensional results. The water heated mold produced parts that had

consistent and acceptable DSC results throughout the length the part; however, the parts

produced from the electrically heated mold had differing crystallinity percentages throughout

any single part.

15 Rev C

3. RESULTS

3.1 Oil versus Pressurized water

The tool used showed a larger flow rate for the oil unit. This is as would be expected with a

rated flow of 100 liters/hour as compared to 60 for the water unit. Temperature differentials for

incoming and outgoing fluid for the oil unit averaged 3.75C. The water unit average 2C.

Both numbers are within the range of typical values for tools flowing at optimum rates. Based

upon this, it is unlikely that either systems efficiency in heat transfer would benefit from

additional flow.

Not charted in the study is that fact that the oil unit averaged a 2.75C differential across the

tool face while the water was unit under 1C.

3.2 Electric Cartridge heaters versus Pressurized water

The design of the electric heating system can be improved with multiple control zones and

faster thermocouple response times. Unfortunately, this would raise energy consumption and

equipment costs. By enlarging the tool passages by 40%, the pressurized water system would

raise the temperature closer to the set point, reduce the variation across the tool, and consume

less energy.

Although not tested for in this test, the cooling cycle can be the longer cycle for molding. The

water unit offers cooling through the same channels as the heating fluid and can be quickly

cycled. The electric cartridge heaters offer no cooling. Cooling systems do exist to compliment

electric heating systems but require an additional chiller and channels to be added to the mold.

Experience with our customers has shown that the cooling cycle is of major concern and is the

longer cycle by a factor of five (5) for thermoplastic composite molding applications.

4. CONCLUSIONS

4.1 Oil versus Pressurized water

Even with 2x the heating capacity (24kW vs. 12kW), 2.8x the cooling capacity (116 kW vs.

41kW), and 1.6x higher rated flow (100 Liters/min. vs. 60 Liters/min.), the oil unit was unable

to reduce heating and cooling times of the mold as compared to the water unit. The oil unit also

consumed 69% more electricity. The most concerning limitation for the oil unit is the large

temperature difference between the TCU and the mold.

4.2 Electric Cartridge heaters versus Pressurized water

Although the tool was not optimized for the water unit, the unit appears to perform better in

every category as compared to electric cartridge heaters.